Re: [tcpm] Feedback request on draft-ietf-tcpm-tcp-security

Lars Eggert <> Mon, 01 March 2010 23:26 UTC

Return-Path: <>
Received: from localhost (localhost []) by (Postfix) with ESMTP id 762E93A8BCA; Mon, 1 Mar 2010 15:26:25 -0800 (PST)
X-Virus-Scanned: amavisd-new at
X-Spam-Flag: NO
X-Spam-Score: -6.151
X-Spam-Status: No, score=-6.151 tagged_above=-999 required=5 tests=[AWL=0.448, BAYES_00=-2.599, RCVD_IN_DNSWL_MED=-4]
Received: from ([]) by localhost ( []) (amavisd-new, port 10024) with ESMTP id ba73HstWkWQx; Mon, 1 Mar 2010 15:26:24 -0800 (PST)
Received: from ( []) by (Postfix) with ESMTP id 538DE3A8BCD; Mon, 1 Mar 2010 15:26:24 -0800 (PST)
Received: from ( []) by (Switch-3.3.3/Switch-3.3.3) with ESMTP id o21NQMUO032190; Mon, 1 Mar 2010 17:26:23 -0600
Received: from ([]) by with Microsoft SMTPSVC(6.0.3790.3959); Tue, 2 Mar 2010 01:26:21 +0200
Received: from ([]) by over TLS secured channel with Microsoft SMTPSVC(6.0.3790.3959); Tue, 2 Mar 2010 01:26:21 +0200
Received: from ( []) by (Switch-3.3.3/Switch-3.3.3) with ESMTP id o21NQJee022432 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NO); Tue, 2 Mar 2010 01:26:20 +0200
X-Virus-Status: Clean
X-Virus-Scanned: clamav-milter 0.95.3 at
Mime-Version: 1.0 (Apple Message framework v1077)
Content-Type: multipart/signed; boundary="Apple-Mail-30--731526454"; protocol="application/pkcs7-signature"; micalg="sha1"
From: Lars Eggert <>
In-Reply-To: <>
Date: Mon, 01 Mar 2010 15:26:05 -0800
Message-Id: <>
References: <>
X-Mailer: Apple Mail (2.1077)
X-Greylist: Sender succeeded SMTP AUTH, not delayed by milter-greylist-4.2.3 ( []); Tue, 02 Mar 2010 01:26:13 +0200 (EET)
X-OriginalArrivalTime: 01 Mar 2010 23:26:21.0775 (UTC) FILETIME=[9A424DF0:01CAB996]
X-Nokia-AV: Clean
Cc: " WG" <>, The IESG <>
Subject: Re: [tcpm] Feedback request on draft-ietf-tcpm-tcp-security
X-Mailman-Version: 2.1.9
Precedence: list
List-Id: TCP Maintenance and Minor Extensions Working Group <>
List-Unsubscribe: <>, <>
List-Archive: <>
List-Post: <>
List-Help: <>
List-Subscribe: <>, <>
X-List-Received-Date: Mon, 01 Mar 2010 23:26:25 -0000


you are raising the serious accusation that the TCPM WG is dysfunctional. It will come as no surprise that I don't share this view.

Yes, the bar for changing TCP is high. TCP is one of the core protocols of the Internet, and the bar for changes *should* be high. I know you follow what is going on in other areas, and neither BGP, DNS, SIP or IP are easily modified. You might not like it, but this is a feature.

We've had several documents recently for which we discussed whether they should be on the Standards Track or not. In some cases, the WG decided for Informational, in other cases, for PS. The chairs called the consensus and the WG agreed.

Are the arguments in TCPM sometimes heated? Absolutely. But that is not a sign of a dysfunctional WG. It just means that the participants feel strongly about the particular technical issues.

Now, let me comment on some of your specific accusations.

On 2010-3-1, at 13:59, wrote:
> Do you really not want to realize that so many folks do not contribute
> any more to TCPM because work in this WG is continually obstructed?

Please back this statement up. Who are the "many folks" who have stopped to contribute because the WG is "obstructed"?

> This is a matter of the experience in interoperable implementations
> being silenced by stereotypous arguments over many years in this WG.

Interoperable implementations are one factor in the IETF standardization process. Rough WG consensus is another. The former does not trump the latter.

> A message sent to this WG at Fri, 19 Feb 2010 15:22:49 -0800
> once more concluded with:
>> I have criticized these documents for proposing changes to standards;
>> I do not have an issue with merely documenting those changes.
> This clearly restates the goal of all this obstruction we had over
> several years and that has caused so many folks to stop active
> participation in this WG.

You again insinuate that "many folks" have left the WG. Again, please back this up. You also imply that there is organized obstruction to TCP changes in the WG. Again, back this accusation up with data.

> That's why this WG does not fulfill its mission and does not perform
> actual maintenance of the protocols it pretends to shepherd.

The WG is publishing documents. There is intense discussion. The documents are of high quality. How exactly does the WG not fulfill its mission?

Yes, the WG does not take on all or even many of the proposals that are brought to it. This is a sign that the WG is taking its stewardship of the TCP protocol seriously.

> I cannot imagine a more concise proof of the diagnose I had placed
> in a recent posting.
> As a mathematician, I can't resist to say:  Q.E.D.   !

As a mathematician, you should know that you have not presented a proof. You have presented an argument.

> It is "the WG", who prohibits this, not _the WG at large_ that has
> been silenced more and more.

There is no "WG at large". The TCPM WG is defined by the individuals that speak up on the TCPM list or in the meetings. Folks who would like to make changes to TCP need to invest the time and effort to convince the community of the value of those changes. Anyone can do so, and many folks do. It is simply false to say that anyone "has been silenced." 

> For me personally, it does not make much sense to waste time and
> energy for further technical and editorial reasoning on documents
> on this list, as long as the frequency of postings iterating the
> same position ad nauseam and the voices of a few folks that happen
> to attend many IETF meetings and get recruited to a WG sessions are
> allowed to redefine the consensus of the WG and to take IETF WG
> procedures ad absurdum.

Nobody "gets recruited" for WG sessions. Like all other IETF WGs, participation is open.

More importantly, you claim that some group of people has "redefined" the WG consensus. This is a serious accusation against the chairs and me as AD. Please provide specific details or retract this accusation.

> Please either shut down the WG now, or (what I would prefer much more)

I'm not shutting the WG down. 

> encourage implementors to return and raise their voice again in the WG,

Implementors have been and continue to be welcome to bring proposals to the WG. I know that several WG participants are actively trying to encourage this. For example, the last time I personally did this was on February 16, when I contacted the proponents of the "TCP thin-stream detection" mechanism that is proposed for inclusion in the Linux kernel to write up their proposal for discussion in the IETF.

(And for the third time, you insinuate that implementors has left the WG, and I'm again asking you to back that statement up.)

> and commit to adhere to the principles Jari Arkko had recalled recently,
> namely that the IETF should much more listen again to running code and
> experience.

Running code is great. So is rough WG consensus. One does not trump the other.